NoSleep

• 

  Example Detection

Inspiration

I wanted to experiment with how software could make our lives safer on the road, or on other critical tasks where sleep deprivation in the operator can be a risk to the lives of hundreds of people.

What it does

The software monitors a camera feed and detects when the person is asleep or about to fall a sleep and alerts them with a loud alarm.

How we built it

The program is built with Python, OpenCV, Dlib, and Yolov5. It uses a combination of landmark detection and AI Object detection to predict when a person is likely asleep.

Challenges we ran into

Training a model for drowsiness detection was challenging in such a short amount of time due to a lack of available datasets.

What's next for NoSleep

The next step for NoSleep is to port it onto mobile devices so that anyone with a phone can simply run the app the while their phone is on their dashboard.

Built With

• dlib
• opencv
• python
• yolov5